ICLR 2026

5,370 papers

Top Research Directions

Research directions most represented by ICLR 2026 papers.

Scaling Language Model Reasoning Beyond Verifiable Rewards 88
Group Relative Policy Optimization for Large Reasoning Models 81
General Purpose Reasoning from Specialized Reinforcement Learning 79
Shifting from Static Benchmarks to Dynamic Evaluation 66
The Decoupling of Visual Perception and Logical Reasoning 51
Mitigating the Inference Latency of Chain of Thought 50
Activation Steering for Inference Time Model Control 41
Preventing Entropy Collapse in Reinforcement Learning 41
Inference Time Compute Scaling for Complex Reasoning 38
Reinforcement Learning with Verifiable Rewards 38
Autonomous Code Refinement via Execution Feedback Loops 38
Mitigating Computational Overthinking in Large Reasoning Models 37
Resolving Superposition with Sparse Autoencoders 34
Role Specialization in Multi-Agent Architectures 34
The Spatial Reasoning Bottleneck in Multimodal Models 33

Top Authors

AuthorPapers
Fei Huang20
Ge Zhang19
Jianye HAO19
Dahua Lin17
Junchi Yan17
Michael M. Bronstein17
Yejin Choi17
James Y. Zou16
James Zou16
Bo Han15
Jingren Zhou15
Min Zhang15
Tat-Seng Chua15
Caiming Xiong14
Dawn Song14
Pengfei Wan14
Yi Yang14
Jiaheng Liu13
Kai Chen13
Pengjun Xie13

Impactful Papers

Papers from ICLR 2026 that shaped research clusters.

Foundational

Papers

Title
DARE-bench: Evaluating Modeling and Instruction Fidelity of LLMs in Data Science
Darwin Gödel Machine: Open-Ended Evolution of Self-Improving Agents
DASH: Deterministic Attention Scheduling for High-throughput Reproducible LLM Training
Data-Aware and Scalable Sensitivity Analysis for Decision Tree Ensembles
Data-Centric Lessons To Improve Speech-Language Pretraining
Dataless Weight Disentanglement in Task Arithmetic via Kronecker-Factored Approximate Curvature
DataMIL: Selecting Data for Robot Imitation Learning with Datamodels
Data Provenance for Image Auto-Regressive Generation
Data Selection for LLM Alignment Using Fine-Grained Preferences
Dataset Color Quantization: A Training-Oriented Framework for Dataset-Level Compression
Dataset Distillation as Pushforward Optimal Quantization
Dataset Distillation for Memorized Data: Soft Labels can Leak Held-Out Teacher Knowledge
Data-to-Energy Stochastic Dynamics
DAVE: A VLM Vision Encoder for Document Understanding and Web Agents
DaVinci: Reinforcing Visual-Structural Syntax in MLLMs for Generalized Scientific Diagram Parsing
DCFold: Efficient Protein Structure Generation with Single Forward Pass
DeAltHDR: Learning HDR Video Reconstruction from Degraded Alternating Exposure Sequences
DEAS: DEtached value learning with Action Sequence for Scalable Offline RL
Death of the Novel(ty): Beyond N-Gram Novelty as a Metric for Textual Creativity
Debiased and Denoised Representation Learning for Incomplete Multi-view Clustering
Debiased Front-Door Learners for Heterogeneous Effects
Debugging Concept Bottleneck Models through Removal and Retraining
DecAlign: Hierarchical Cross-Modal Alignment for Decoupled Multimodal Representation Learning
Decentralized Attention Fails Centralized Signals: Rethinking Transformers for Medical Time Series
Decentralized Nonconvex Optimization under Heavy-Tailed Noise: Normalization and Optimal Convergence
Decision Aggregation under Quantal Response
Decision-Theoretic Approaches for Improved Learning-Augmented Algorithms
DeCo-DETR: Decoupled Cognition DETR for efficient Open-Vocabulary Object Detection
Decoding Dynamic Visual Experience from Calcium Imaging via Cell-Pattern-Aware Pretraining
Decoding Open-Ended Information Seeking Goals from Eye Movements in Reading
DecompGAIL: Learning Realistic Traffic Behaviors with Decomposed Multi-Agent Generative Adversarial Imitation Learning
Decomposed Attention Fusion in MLLMs for Training-free Video Reasoning Segmentation
Decomposing LLM Computation with Jets
Decomposing Representation Space into Interpretable Subspaces with Unsupervised Learning
Decomposition of Concept-Level Rules in Visual Scenes
Deconstructing Guidance: A Semantic Hierarchy for Precise Diffusion Model Editing
Deconstructing Positional Information: From Attention Logits to Training Biases
Decoupled DMD: CFG Augmentation as the Spear, Distribution Matching as the Shield
Decoupled MeanFlow: Turning Flow Models into Flow Maps for Accelerated Sampling
Decoupled Q-Chunking
Decoupling Dynamical Richness from Representation Learning: Towards Practical Measurement
Decoupling Positional and Symbolic Attention in Transformers
Decoupling Primitive with Experts: Dynamic Feature Alignment for Compositional Zero-Shot Learning
Decoupling the Class Label and the Target Concept in Machine Unlearning
DeepAFL: Deep Analytic Federated Learning
DeepCompress: A Dual Reward Strategy for Dynamically Exploring and Compressing Reasoning Chains
DeepEyes: Incentivizing "Thinking with Images" via Reinforcement Learning
DeepEyesV2: Toward Agentic Multimodal Model
Deep FlexQP: Accelerated Nonlinear Programming via Deep Unfolding
DeepFRC: An End-to-End Deep Learning Model for Functional Registration and Classification

« prev Page 21/108 next »